System and method for early warning natural disasters influencing electric power facilities based on digital employees

Abstract

The invention discloses an early warning system and method for natural disasters influencing electric power facilities based on digital employees. The early warning system comprises an automatic disaster weather identification unit, an information intercepting and sending unit and an early warning information reporting unit. The disaster weather automatic identification unit automatically acquires and identifies weather characters and images through digital employees based on an RPA method in combination with OCR, image identification, natural language processing and template matching methods, and compares and judges whether the weather information accords with natural disaster weather information influencing electric power facilities; the information intercepting and sending unit intercepts disastrous weather early warning characters and images which accord with natural disaster weather which affects the electric power facilities after classification judgment through digital employees, and sends the early warning characters and images to the terminal by adopting an RPA method; the early warning information reporting unit processes disaster weather early warning characters and images through digital employees to generate reporting information.

Description

Technical Field

[0001] The present invention relates to the technical field of disaster early warning, and more specifically, to a natural disaster early warning system for power facilities based on digital employees. Background Art

[0002] Abnormal natural disaster weather can pose a hazard to local power facilities and may also affect the power consumption perception of customers. Timely obtaining natural disaster weather that affects power facilities can alert staff to take countermeasures in advance, strengthen the on-duty staff, and also report to the staff of the State Grid to reduce service risks in the event of disaster weather and prevent the expansion of incidents. Therefore, it is of great significance to develop an automatic early warning model for natural disaster weather that affects power facilities, use digital employees to replace manual operations, quickly and accurately obtain disaster weather early warning information without omission in the first time, reduce the impact of disaster weather on power production, ensure the safe and stable operation of the power grid, and promote cost reduction and efficiency improvement of enterprises and increase economic benefits.

[0003] To ensure the safe and stable operation of power facilities within the State Grid and timely avoid the hazards that abnormal natural disaster weather may pose to local power facilities and the possible impacts on customers' power consumption perception, currently, on-duty personnel need to query weather forecasts and early warning information from the China Weather Network and the National Emergency Early Warning Information Release Network, and intercept text messages to send to relevant staff to alert them to take countermeasures and increase on-duty standby. However, this method of querying weather information and sending text message warnings has the following disadvantages:

[0004] (1) Large number of monitoring tasks. It is necessary to continuously monitor the information on the weather early warning website within 24 hours to obtain natural disaster weather in the urban area and 17 counties. The monitoring time is long, the area is wide, the tasks are heavy, and it is time-consuming and laborious.

[0005] (2) Low work efficiency: Staff need to monitor and query information on natural disaster weather of 14 types that affect power facilities in multiple regions of the city and county, and compare whether they meet 3 different disaster weather levels, which is prone to missed checks, wrong checks, and mischecks.

[0006] (3) Slow early warning time: Natural disaster weather is highly harmful to power production. It is necessary to capture information in the first time, then edit text messages with different meteorological information contents, and select and send them to the staff in the designated area, making it difficult to conduct rapid information early warning and reporting in real time.

[0007] (4) High work cost: Manually monitoring natural disaster weather information continuously requires a large amount of manpower. Sending a large number of disaster weather text messages has a high dependence on manual labor, increasing economic costs and being unfavorable for work development.

[0008] Publication number: CN111966746A, titled "A Computer-Implemented Method for Monitoring Meteorological Disaster Prevention and Mitigation Processes", including: Lane Receiving Step: Receiving meteorological data using at least one meteorological station data live lane, and receiving event data using at least one event data lane; Lane Data Processing Step: Parsing the time, event location, and event type of the event data, and parsing the time and data volume values in the meteorological data of the meteorological stations around the event location; Lane Interface Display Step: Generating an event thumbnail based on the time and event type of the event data and displaying it on a discrete lane unit of the discrete lane, and generating a curve or bar chart based on the time and data volume values of the meteorological data and displaying it on the meteorological station data live lane.

[0009] Publication number: CN106097661A, titled "A Method and System for Early Warning of Power Equipment Risks Considering Meteorological Disasters", including the following steps: Obtaining current meteorological disaster early warning information; Obtaining current power grid security hidden danger information; Judging whether there is a meteorological disaster early warning in the area where the power grid security hidden danger is located; Judging the type and degree of the early warning in the area where the power grid security hidden danger is located; Issuing a security hidden danger early warning.

[0010] In summary, the existing power systems have scattered sources for obtaining natural disaster-related weather information and lack effective integration, resulting in insufficient comprehensiveness of disaster information; traditional manual screening and data analysis affect the accuracy of information; the time lag in the information transmission and response processes affects the rapidity of emergency response; the number of employees for seasonal concentrated abnormal weather early warning and reporting is limited, affecting the safe operation of power facilities in extreme situations.

[0011] Therefore, how to provide a method to quickly and efficiently obtain natural disaster-related weather that affects power facilities, enable early warning personnel to take countermeasures in advance, reduce the impact of disaster weather on customer electricity consumption, report to the State Grid business support system, and ensure the safe and stable operation of the regional power grid is an urgent problem to be solved by those skilled in the art. Summary of the Invention

[0012] In view of this, the present invention provides a natural disaster early warning system for power facilities based on digital employees, aiming to solve the technical problems that the existing power systems have scattered sources for obtaining natural disaster-related weather information and lack effective integration, resulting in insufficient comprehensiveness of disaster information; traditional manual screening and data analysis affect the accuracy of information; the time lag in the information transmission and response processes affects the rapidity of emergency response; the number of employees for seasonal concentrated abnormal weather early warning and reporting is limited, affecting the safe operation of power facilities in extreme situations.

[0013] To achieve the above object, the present invention adopts the following technical solutions:

[0014] A natural disaster early warning system for power facilities based on digital employees, comprising: a disaster weather automatic recognition unit, an information interception and sending unit, and an early warning information reporting unit;

[0015] The disaster weather automatic recognition unit automatically obtains and recognizes meteorological texts and images through digital employees based on the RPA method combined with OCR, image recognition, natural language processing, and template matching methods, and compares and determines the weather information of natural disasters that meet the conditions for affecting power facilities;

[0016] The information interception and sending unit intercepts the warning texts and images of disaster weather after the classification determination of natural disasters that meet the conditions for affecting power facilities through digital employees, and uses the RPA method to send the warning texts and images to the terminal;

[0017] The early warning information reporting unit processes the warning texts and images of disaster weather through digital employees to generate reporting information.

[0018] Further, the disaster weather automatic recognition unit specifically includes an information monitoring unit, an information query and capture unit, a weather classification comparison unit, and an early warning classification unit;

[0019] The information monitoring unit, through digital employees, monitors early warning information in real time, automatically generates a monitoring log, records the execution times, time nodes, and abnormal situations, and forms a data monitoring report;

[0020] The information query and capture unit, based on the administrative region planning, automatically enters the administrative regions of preset areas by digital employees and obtains the early warning information corresponding to the administrative regions;

[0021] The weather classification comparison unit, based on a preset natural disaster classification table, uses the template matching method and natural language processing method to perform natural disaster weather classification recognition;

[0022] The early warning classification unit, based on the disaster level standard, classifies the recognized natural disaster weather.

[0023] Further, the information query and capture unit specifically is:

[0024] Batch input the names of administrative regions, and loop to execute the area switching query instruction;

[0025] Batch search and capture natural disaster weather early warning information, and based on the RPA method, execute multiple search and capture processes in parallel while performing area information comparison.

[0026] Further, the weather classification comparison unit specifically is:

[0027] Based on the preset classification of natural disasters affecting power facilities, the digital employee uses template matching technology and natural language processing technology to automatically classify and identify natural disaster weather, compare natural disaster weather, and extract key information in the comparison results through the RPA method, including disaster type, occurrence time, and affected area. The captured data is compared with the set classification template, and irrelevant information is filtered out to identify disaster weather affecting power facilities.

[0028] Furthermore, the information interception and sending unit includes an information extraction unit and an information sending unit;

[0029] For the information interception unit, the digital employee, based on RPA technology and natural language processing technology, identifies and obtains weather warning information related to power facilities, and automatically filters out eligible disaster weather warning texts according to the types of natural disasters affecting power facilities. The digital employee automatically identifies and intercepts image information related to disasters, judges whether the image meets the safety warning requirements of power facilities according to the warning criteria, and automatically intercepts and saves the compliant images.

[0030] For the information sending unit, it automatically obtains eligible disaster weather warning texts and image information that meet the safety warning requirements of power facilities and sends them to the terminal.

[0031] Furthermore, the warning information reporting unit includes an automatic login unit, an information entry unit, and an attachment information upload unit;

[0032] For the automatic login unit, the digital employee automatically logs in to the reporting system, locates to the reporting interface, and detects whether the interface has finished loading;

[0033] For the information entry unit, the digital employee matches the disaster category based on the intercepted disaster weather warning text, enters the key information, and verifies the integrity of the entered key information;

[0034] For the attachment information upload unit, the digital employee uploads the intercepted disaster weather warning image as supporting attachment information to the business support system.

[0035] Furthermore, the specific key information is as follows:

[0036] Impact time: The digital employee extracts the disaster occurrence time or prediction period mentioned in the disaster weather warning text and automatically enters it into the impact time field.

[0037] Affected area: The digital employee selects the corresponding affected area option according to the administrative region information in the disaster warning content.

[0038] Impact level: The digital employee automatically enters the corresponding impact level according to the warning level.

[0039] Furthermore, the attachment information uploading unit further includes:

[0040] During the file uploading process, the file format, size, and uploading progress are automatically verified. After the uploading is completed, the task execution log is automatically recorded, including time nodes, image file names, and uploading status.

[0041] A natural disaster early warning method for power facilities based on digital employees includes the following steps:

[0042] The digital employee automatically monitors disaster weather information, automatically intercepts and enters the administrative region name, searches for natural disaster weather information one by one, compares the natural disaster weather information with the weather information and disaster levels, and automatically identifies disaster weather;

[0043] Identifies and automatically intercepts compliant disaster weather warning images and texts, generates warning reporting information with consistent requirements, and sends the warning reporting information to the mobile terminal;

[0044] After obtaining the warning information, it automatically reports to the system. According to the intercepted disaster weather warning text, it selects the corresponding natural disaster category in the reporting interface and enters the impact time, impact area, impact classification, and impact level. The disaster weather warning image is used as the supporting attachment information and is automatically uploaded to the business support system.

[0045] Furthermore, the digital employee automatically monitoring disaster weather information further includes:

[0046] The digital employee monitors the effectiveness of the disaster weather information acquisition end. If the acquisition end is invalid, it refreshes and re-acquires, and verifies the execution status of the information query instruction after a preset time.

[0047] The above technical solution at least includes the following technical effects:

[0048] The natural disaster automatic early warning model for power facilities involves three links: automatic identification of disaster weather, information interception and sending, and warning information reporting. By introducing the concept of digital employees for integrated design, development, and testing implementation, information interaction between the Internet web page and WeChat software can be achieved. It can not only improve the efficiency, accuracy, and real-time nature of manual monitoring, but also avoid the investment in new software and the cost of sending text messages, resulting in significant cost reduction, efficiency improvement, and economic benefits.

[0049] (1) Adopting the automatic monitoring mode, it automatically executes multi-region, multi-type, and multi-level monitoring tasks without manual intervention, improving the monitoring efficiency. This efficient monitoring mechanism can predict in advance and quickly formulate solutions, significantly shortening the power interruption time and reducing the economic losses caused by power outages;

[0050] (2) High accuracy in obtaining information. Automatic monitoring and information comparison are carried out according to preset rules, avoiding omissions and errors caused by manual operations, and reducing some incidents of power facility damage caused by false alarms and missed reports.

[0051] (3) The monitoring results are fed back in real time. The monitoring results are generated in real time, providing timely and accurate meteorological information in text and pictures, which can respond in a timely manner and report quickly, reducing the impact on residents' lives and industrial production, and enhancing users' satisfaction and the power company's reputation in the public.

[0052] (4) Cost reduction, efficiency increase, speed improvement and quality enhancement: Digital employee operations replace manual verification, reducing the cost of monitoring, warning and reporting, improving the quality and efficiency of the enterprise, and having the value of promoting the whole region of provinces, cities and counties and good cost competitiveness. Description of the Drawings

[0053] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only the embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained according to the provided drawings without creative efforts.

[0054] Figure 1 It is a schematic structural diagram of a natural disaster weather warning model process of the present invention.

[0055] Figure 2 It is a schematic structural diagram of a process for real-time monitoring and identifying natural disaster weather affecting power facilities of the present invention.

[0056] Figure 3 It is a schematic structural diagram of a process for automatic warning and reporting of natural disaster weather of the present invention. Detailed Embodiments

[0057] The following clearly and completely describes the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0058] See the appendix Figures 1 - 3 In the embodiments of the present invention, a natural disaster warning system for power facilities based on digital employees is disclosed, including: a disaster weather automatic recognition unit, an information intercepting and sending unit, and a warning information reporting unit;

[0059] The automatic disaster weather recognition unit, based on the RPA method combined with OCR, image recognition, natural language processing and template matching methods, automatically obtains and recognizes meteorological texts and images through digital employees, and compares and determines the natural disaster weather information that affects power facilities;

[0060] The automatic disaster weather recognition unit specifically includes an information monitoring unit, an information query and scraping unit, a weather classification and comparison unit, and a warning classification unit;

[0061] The information monitoring unit, the digital employee monitors the warning information in real time, and automatically generates a monitoring log, records the execution times, time nodes and abnormal situations, and forms a data monitoring report;

[0062] The information query and scraping unit, the digital employee, based on the administrative region planning, automatically enters the administrative regions of the preset regions, and obtains the warning information corresponding to the administrative regions; specifically: batch enter the names of administrative regions, and loop execute the region switching query instruction; batch search and scrape the natural disaster weather warning information, and based on the RPA method, execute multiple search and scraping processes in parallel, and at the same time conduct regional information comparison;

[0063] The weather classification and comparison unit, the digital employee, based on the preset natural disaster classification table, uses the template matching method and the natural language processing method to conduct natural disaster weather classification and recognition; specifically: the digital employee, based on the preset natural disaster classification that affects power facilities, uses the template matching technology and the natural language processing technology to automatically conduct the classification and recognition of natural disaster weather, compares the natural disaster weather, and extracts the key information in the comparison result through the RPA method, including the disaster type, occurrence time and influence range, compares the grabbed data with the set classification template, and filters out irrelevant information to identify the disaster weather that affects power facilities

[0064] The warning classification unit, the digital employee, based on the disaster level standard, divides the recognized natural disaster weather into levels.

[0065] The information intercepting and sending unit, intercepts the disaster weather warning texts and images after the classification determination of the natural disaster weather that affects power facilities through the digital employee, and uses the RPA method to send the warning texts and images to the terminal; the information intercepting and sending unit includes an information extraction unit and an information sending unit;

[0066] An information interception unit. Based on RPA technology and natural language processing technology, the digital employee identifies and obtains weather warning information related to power facilities, and automatically filters out eligible disaster weather warning texts according to the types of natural disasters affecting the power facilities. The digital employee automatically identifies and intercepts image information related to the disaster, determines whether the image meets the safety warning requirements of the power facilities according to the warning criteria, and automatically intercepts and saves the compliant images. An information sending unit automatically obtains eligible disaster weather warning texts and image information that meet the safety warning requirements of the power facilities, and sends them to the terminal.

[0067] A warning information filing unit processes the disaster weather warning texts and images through the digital employee to generate filing information. The warning information filing unit includes an automatic login unit, an information entry unit, and an attachment information upload unit. The automatic login unit: The digital employee automatically logs in to the filing system, locates to the filing interface, and detects whether the interface has finished loading. The information entry unit: Based on the intercepted disaster weather warning texts, the digital employee matches the disaster categories, enters the key information, and verifies the integrity of the entered key information. The attachment information upload unit: The digital employee takes the intercepted disaster weather warning images as supporting attachment information and uploads them to the business support system. During the file upload process, it automatically verifies the file format, size, and upload progress, and automatically records the task execution log after upload, including time nodes, image file names, and upload status.

[0068] The key information specifically includes:

[0069] Influence time: The digital employee extracts the disaster occurrence time or predicted period mentioned in the disaster weather warning text and automatically enters it into the influence time field.

[0070] Influence area: The digital employee selects the corresponding influence area option according to the administrative region information in the disaster warning content.

[0071] Influence level: The digital employee automatically enters the corresponding influence level according to the warning level.

[0072] A method for natural disaster warning affecting power facilities based on a digital employee includes the following steps:

[0073] The digital employee automatically monitors disaster weather information, automatically intercepts and enters the administrative region names, searches for natural disaster weather information one by one, compares the natural disaster weather information with the weather information and disaster levels, and automatically identifies disaster weather; identifies and automatically intercepts compliant disaster weather warning images and texts, generates warning reporting information with consistent requirements, and sends the warning reporting information to the mobile terminal; after obtaining the warning information, it automatically reports to the system, and according to the intercepted disaster weather warning text, selects the corresponding natural disaster type in the reporting interface to enter the impact time, impact area, impact classification, and impact level, and takes the disaster weather warning image as the supporting attachment information and automatically uploads it to the business support system. The digital employee monitors the effectiveness of the disaster weather information acquisition terminal. If the acquisition terminal is invalid, it refreshes and re-obtains, and verifies the execution status of the information query instruction after a preset time.

[0074] Taking Shijiazhuang City, Hebei Province and its 17 subordinate counties as an example, the technical solution of this application will be described in detail:

[0075] The disaster weather automatic recognition unit uses RPA technology combined with OCR, image recognition, natural language processing (NLP) and template matching technology, introduces the concept of digital employees, automatically recognizes meteorological texts and images in the National Emergency Warning Information Release Network, and compares and determines natural disaster weather information that affects power facilities, including regional information, Shijiazhuang City, Hebei Province and its 17 subordinate counties; category information, 14 types of weather such as lightning, rainstorm, high temperature, etc.; level information, 3 disaster levels of yellow warning, orange warning, and red warning.

[0076] Step (1): 24-hour automatic monitoring and on-duty verification of digital employees

[0077] The digital employee performs 7x24-hour real-time monitoring of the National Emergency Warning Information Release Network through a preset process script, simulating the continuous monitoring work of manual duty.

[0078] Every 1 hour, the digital employee automatically verifies the execution status of the information query instruction to ensure the normal operation of data capture and monitoring tasks. Every 2 hours, the digital employee executes the "on-duty check-in" task, sends a check-in request and performs verification through a predetermined time node to prove the continuous on-duty status of the digital employee and ensure the efficient and stable operation of the system.

[0079] The digital employee can automatically generate a monitoring log, record the execution times, time nodes and abnormal situations, and form a traceable data monitoring report.

[0080] Step (2): Automatic interception and query of multi-dimensional information of administrative regions

[0081] According to the government's administrative regional planning, the digital employee is set by a script to automatically enter the names of 17 counties (such as Zhengding, Luquan, Gaocheng, Luancheng, etc.) and the main urban areas under the jurisdiction of Shijiazhuang City, Hebei Province, a total of 18 administrative regions. The system queries one by one in the National Emergency Warning Information Release Network and automatically executes the following tasks:

[0082] Automatically batch enter the names of administrative regions and loop execute the area switching query command.

[0083] Batch search and capture natural disaster weather warning information to ensure the comprehensiveness of area coverage and the efficiency of information collection.

[0084] Technical expansion: RPA can execute multiple query processes in parallel, compare area information simultaneously, shorten the information capture time, and ensure real-time and comprehensiveness.

[0085] Step (3): Automatic classification and intelligent comparison of natural disaster weather

[0086] Based on the classification of natural disasters affecting power facilities stipulated by the State Grid, the digital employee uses template matching technology and natural language processing (NLP) technology to automatically classify and identify natural disaster weather. It mainly includes:

[0087] Automatically and intelligently compare 14 types of natural disaster weather such as lightning, rainstorm, high temperature, hail, etc., and extract key information in the comparison results through RPA, such as disaster type, occurrence time, affected area, etc.

[0088] Intelligently compare the captured data with the set classification template, filter out irrelevant information, and accurately identify the disaster weather affecting power facilities.

[0089] Through the NLP algorithm, the digital employee can automatically analyze the key sentences in the text description, distinguish the category and intensity of natural disaster weather, and improve the recognition accuracy.

[0090] Step (4): Automatic comparison of disaster levels and early warning classification

[0091] The system divides the disaster levels into three levels: yellow warning, orange warning, and red warning according to the State Grid disaster level standard, and automatically compares the weather warning level standard for the identified natural disaster weather through the RPA system. According to the comparison results of the disaster levels, generate an early warning report, and automatically mark key information such as disaster level, affected area, and duration.

[0092] The digital employee can automatically generate an early warning classification statistical chart to visually display the disaster level distribution of each administrative region, facilitating quick decision-making by managers.

[0093] Table 1 Classification of natural disaster weather affecting power facilities

[0094]

[0095]

[0096] The information intercepting and sending unit obtains weather data from the National Emergency Warning Information Release Network through digital employees, performs automated processing, automatically intercepts the warning texts and images of disastrous weather after the classification judgment of natural disaster weather types that affect power facilities through OCR and NLP technologies, triggers a WeChat mini-program using RPA technology, copies and pastes them into the WeChat dialog box, and sends the text and image information to power staff, so as to obtain the warning information of natural disaster weather affecting power facilities accurately and without omission in the first time.

[0097] Step (1): Automatically intercept the warning texts of disastrous weather after the classification judgment of natural disaster weather types that affect power facilities.

[0098] The digital employee automatically identifies the weather warning information related to power facilities from the meteorological bureau or other disaster information release platforms through RPA and natural language processing (NLP) technologies. The system automatically filters out the warning texts of eligible disastrous weather according to the types of natural disasters that power facilities may be affected by (such as lightning, heavy rain, typhoon, etc.).

[0099] The digital employee sets rules through RPA to automatically extract key information in the warning text, such as disaster type, affected area, warning level, occurrence time, etc., to ensure the accuracy of the extracted text information.

[0100] Step (2): Automatically intercept the warning images of disastrous weather after the classification judgment of natural disaster weather types that affect power facilities

[0101] In many cases, the warning information of disastrous weather is released in the form of images or charts. To ensure that power facilities can comprehensively understand the warning situation, the RPA digital employee will automatically identify and intercept the image information related to the disaster. The system analyzes the key information in the image through image recognition technologies (such as OCR and image processing algorithms), including but not limited to disaster path maps, intensity distribution maps, radar maps, etc.

[0102] RPA judges whether the image meets the safety warning requirements of power facilities according to the warning standards and automatically intercepts and saves it.

[0103] Step (3): Automatically trigger the WeChat mini-program through an instruction, and copy and paste the intercepted text and image information into the WeChat dialog box.

[0104] Once the text and image information of disastrous weather are intercepted, the digital employee RPA will automatically trigger the WeChat mini-program through an instruction to achieve automatic operation:

[0105] RPA runs in the background, automatically opens the WeChat mini-program through set instructions, and pastes the intercepted text and image information. Without manual intervention, the system automatically inputs the data into the WeChat dialog box to ensure the smoothness of information transmission.

[0106] Step (4): Send the intercepted text and image information to the designated power staff through WeChat messages to ensure accurate and complete mobile information warning in the first time.

[0107] After completing the input in the WeChat dialog box, the RPA digital employee will automatically send the warning information to the designated power staff. Through instant message push, ensure that the information is quickly and accurately transmitted to the mobile phones of relevant power personnel. The system automatically identifies the recipient list according to the preset rules and automatically selects the corresponding personnel for information push according to the urgency. The WeChat mini-program synchronously sends the text and image information to the power staff to ensure no information omission and achieve the warning effect in the first time.

[0108] The warning information reporting unit, based on RPA technology, automatically generates reporting information from the processed disaster weather warning information. In the State Grid business support system, enter the natural disaster weather affecting power facilities into the system by indicating the impact time, impact area, impact classification, impact level, etc., and automatically upload screenshots of the natural disaster weather affecting power facilities to report to the State Grid staff, reducing the service risk in case of disaster weather, avoiding the expansion of incidents, and supporting the risk prevention of power supply services in Shijiazhuang City, Hebei Province and its 17 subordinate counties.

[0109] Step (1): Automatically log in to the State Grid business support system through instructions and open the important service matter reporting interface.

[0110] Automated login: The RPA digital employee automatically simulates manual operations through a preset process script, inputs credentials such as user names and passwords, and securely logs in to the State Grid business support system.

[0111] Automatic interface navigation: The system automatically locates to the "important service matter reporting" interface, and realizes the rapid loading and switching of the system interface without manual intervention.

[0112] Execution verification: RPA automatically checks whether the page loading is completed to ensure that the system successfully enters the target interface and guarantees the stable operation of the process.

[0113] Step (2): According to the intercepted disaster weather warning text, select the corresponding natural disaster category and enter the impact time, impact area, impact classification, and impact level.

[0114] The digital employee RPA combines the text information of disaster weather warnings intercepted by the front-end system, and through intelligent parsing and classification, automatically completes the entry of reporting information:

[0115] 1. Intelligent matching of disaster categories

[0116] RPA automatically identifies the warning types in the text, such as 14 types of natural disasters including "lightning", "rainstorm", "high temperature", etc., and matches the corresponding "impact classification" options.

[0117] 2. Automatic entry of key information

[0118] Impact time: The system extracts the disaster occurrence time or predicted period mentioned in the text and automatically enters it into the "impact time" field.

[0119] Impact area: RPA selects the corresponding "impact area" option according to the administrative region information (such as the names of counties, cities, and districts) in the disaster warning content.

[0120] Impact level: According to the warning level (such as yellow, orange, red warnings), the system automatically enters the corresponding "impact level".

[0121] 3. Verification of entry integrity

[0122] RPA automatically checks the integrity of the entered content through preset rules to ensure that each field is accurately filled and prevent omissions.

[0123] Step (3): According to the intercepted disaster weather warning image, as the attached information for support, it is automatically uploaded into the business support system

[0124] After completing the entry of text information, the RPA digital employee further automatically processes the warning image information and uploads it as attached supporting materials:

[0125] 1. Automatic attachment matching

[0126] The system identifies the image file intercepted by the front-end and automatically associates it with the current reporting task to ensure the accurate matching of image information.

[0127] 2. Automatic upload operation

[0128] RPA simulates mouse and keyboard operations, automatically clicks the "attachment upload" function, and executes the upload task of the image file. During the file upload process, the system automatically verifies the file format, size, and upload progress to ensure that the attachment meets the system requirements and is successfully uploaded.

[0129] 3. Feedback results and log records

[0130] After the system finishes uploading, it automatically records the task execution log, including time nodes, image file names, and upload status, to achieve traceability and transparent management of the reporting operation.

[0131] The present application has the following beneficial effects:

[0132] First, it adopts an automatic monitoring mode, automatically executes multi-region, multi-type, and multi-level monitoring tasks without manual intervention, improving the monitoring efficiency. This efficient monitoring mechanism can predict in advance and quickly formulate solutions, which can significantly shorten the power interruption time and reduce the economic losses caused by power outages;

[0133] Second, the information obtained is highly accurate. Automatic monitoring and information comparison are carried out according to preset rules, avoiding omissions and errors caused by human operations, and reducing some incidents of power facility damage caused by false alarms and missed reports.

[0134] Third, the monitoring results are fed back in real time. The monitoring results are generated in real time, providing timely and accurate meteorological information in text and pictures, which can respond in a timely manner and report quickly, reducing the impact on residents' lives and industrial production, and enhancing user satisfaction and the power company's reputation in the public.

[0135] Fourth, it reduces costs, increases efficiency, speeds up, and improves quality: Digital employee operations replace manual verification, reducing the cost of monitoring, warning, and reporting, improving the quality and efficiency of the enterprise, and having the value of being promoted across the whole regions of provinces, cities, and counties and having good cost competitiveness.

[0136] In this specification, each embodiment is described in a progressive manner. The key point of each embodiment is to illustrate the differences from other embodiments. The same or similar parts among the embodiments can be referred to each other.

[0137] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present invention. Various modifications to these embodiments will be obvious to those skilled in the art. The general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but will be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A natural disaster early warning system for power facilities based on digital employees, characterized in that: include: Disaster weather automatic identification unit, information interception and sending unit, early warning information reporting unit; The disaster weather automatic identification unit is used to automatically obtain and identify meteorological text and images through digital employees based on the RPA method combined with OCR, image recognition, natural language processing and template matching methods, and compare and determine the natural disaster weather information that affects power facilities; The information interception and sending unit is used to intercept the disaster weather warning text and image that meets the classification and determination of natural disaster weather affecting power facilities through digital employees, and send the warning text and image to the terminal using the RPA method; The warning information reporting unit is used to process the disaster weather warning text and image through digital employees to generate reporting information.

2. According to claim 1, a natural disaster early warning system for power facilities based on digital employees is characterized in that: The disaster weather automatic identification unit specifically includes an information monitoring unit, an information query and capture unit, a weather classification comparison unit, and an early warning classification unit; The information monitoring unit, the digital employee, monitors the early warning information in real time and automatically generates a monitoring log, records the number of executions, time nodes and abnormal conditions, and forms a data monitoring report; The information query and capture unit, the digital employee, automatically enters the administrative area of ​​the preset area based on the administrative area planning, and obtains the warning information corresponding to the administrative area; The weather classification comparison unit, the digital employee, performs natural disaster weather classification identification based on a preset natural disaster classification table through a template matching method and a natural language processing method; The early warning classification unit, the digital employee, classifies the identified natural disaster weather based on the disaster level standard.

3. A natural disaster early warning system for power facilities based on digital employees according to claim 2, characterized in that: The information query and capture unit is specifically: Batch admit administrative area names and execute area switching query instructions cyclically; Batch search and crawl natural disaster weather warning information, execute multiple search and crawl processes in parallel based on the RPA method, and compare regional information at the same time.

4. The natural disaster early warning system for power facilities based on digital employees according to claim 2 is characterized in that: The weather classification comparison unit is specifically: Based on the preset classification of natural disasters that affect power facilities, digital employees use template matching technology and natural language processing technology to automatically classify and identify natural disaster weather, compare natural disaster weather, and extract key information from the comparison results through RPA methods, including disaster type, occurrence time and impact range, compare the captured data with the set classification template, filter out irrelevant information, and identify disaster weather that affects power facilities.

5. The natural disaster early warning system for power facilities based on digital employees according to claim 1 is characterized in that: The information interception and sending unit includes an information extraction unit and an information sending unit; The information interception unit, the digital employee, identifies and obtains weather warning information related to power facilities based on RPA technology and natural language processing technology, and automatically screens out qualified disaster weather warning texts according to the type of natural disasters affected by the power facilities; the digital employee automatically identifies and intercepts image information related to the disaster, and determines whether the image meets the safety warning requirements of the power facilities according to the warning standards, and automatically intercepts and saves the qualified images; The information sending unit automatically obtains the disastrous weather warning text that meets the conditions and the image information that meets the safety warning requirements of the power facilities, and sends them to the terminal.

6. The natural disaster early warning system for power facilities based on digital employees according to claim 1 is characterized in that: The warning information reporting unit includes an automatic login unit, an information input unit and an attachment information uploading unit; The automatic login unit is used to automatically log in to the reporting system through the digital employee, locate the reporting interface, and detect whether the interface has been loaded; Information entry unit: Digital employees match disaster categories based on intercepted disastrous weather warning texts, enter key information, and verify the integrity of the entered key information; The attachment information uploading unit, in which digital employees upload the captured disastrous weather warning images as supporting attachment information to the business support system.

7. A natural disaster early warning system for power facilities based on digital employees according to claim 6, characterized in that: The key information is specifically: Impact time: Digital employees extract the disaster occurrence time or forecast period mentioned in the disastrous weather warning text and automatically enter it into the impact time field; Impact area: Digital employees select the corresponding impact area option based on the administrative area information in the disaster warning content; Impact level: Digital employees automatically enter the corresponding impact level based on the warning level.

8. The natural disaster early warning system for power facilities based on digital employees according to claim 6 is characterized in that: The attachment information uploading unit also includes: During the file upload process, the file format, size and upload progress are automatically verified. After the upload is completed, the task execution log is automatically recorded, including the time node, image file name and upload status.

9. A method for early warning of natural disasters affecting power facilities based on digital employees, applicable to any one of the early warning systems for natural disasters affecting power facilities based on digital employees as described in claims 1-8, characterized in that: The following steps are involved: Digital employees automatically monitor disaster weather information, automatically capture and enter administrative area names, search for natural disaster weather information one by one, compare natural disaster weather information with weather information and disaster levels, and automatically identify disaster weather; Identify and automatically capture the severe weather warning images and texts that meet the requirements, generate warning and preparedness information that meets the requirements, and send the warning and preparedness information to the mobile terminal; After obtaining the warning information, it is automatically reported to the system. According to the intercepted disastrous weather warning text, select the corresponding natural disaster type in the reporting interface, enter the impact time, impact area, impact classification, and impact level, and use the disastrous weather warning image as supporting attachment information, which is automatically uploaded to the business support system.

10. A method for early warning of natural disasters affecting power facilities based on digital employees according to claim 9, characterized in that: The digital workforce automatically monitors severe weather information and also includes: The digital employee monitors the validity of the disaster weather information acquisition terminal. If the acquisition terminal is invalid, it will be refreshed and re-acquired, and the execution of the information query instruction will be verified after a preset time.

Images